New specimens of picromomyids (Plesiadapiformes, Primates) with description of a new species of Alveojunctus

11 Pages·2002·10.6 MB·English

by  Mary T Silcox

ANNALS OF CARNEGIE MUSEUM
VoL. 71, Number 1, Pp. 1-11 26 February 2002
NEW SPECIMENS OF PICROMOMYIDS
(PLESIADAPIFORMES, PRIMATES) WITH DESCRIPTION OF A NEW
SPECIES OF ALVEOJUNCTUS
Mary T Silcox^
Kenneth D. Rose^
Research Associate, Section ofVertebrate Paleontology
Stephen L. Walsh^
Abstract
Two new specimens attributable to the plesiadapiformfamily Picromomyidae aredocumented. One,
a lower molar, represents a new species here named Alveojimctus bowni. This species is the largest
and mostderivedpicromomyidcurrently known, with anextremely simplifiedmolarmorphology. The
second new specimen is a fragmentary lowerjaw attributed to Picromomyspetersonorum; the speci-
men includes the first complete lower central incisor known for a picromomyid. This tooth shows
none of the derived features seen in microsyopid IjS, providing no support for a tie to this family.
Instead, it is most similar to I, in Tinimomys graybulliensis, supporting a close relationship between
Picromomyidae and Micromomyidae.
Key Words; Incisors, Picromomyidae, Micromomyidae, Plesiadapiformes, Primates, Teeth
Introduction
In 1996 Rose and Bown recognized a new family of plesiadapiform based on
the discovery of associated right and left dentaries representing a new species,
Picromomys petersonorum. This species is characterized by an enlarged P4 with
a wide, oddly flattened talonid and unusual, cusp-like, mesiobuccal expansions
on the trigonids ofMj.2. P. petersonorum is also exceptional as one ofthe smallest
primate-like animals ever described. The recent description of the “smallest pri-
mate” by Gebo et al. (2000) does not discuss Picromomys, presumably because
the authors do not consider plesiadapiforms to be primates. A comprehensive
phylogenetic analysis of dental, cranial, and postcranial features of a wide diver-
sity of plesiadapiforms, euprimates, and other archontans indicates that plesia-
dapiforms are most appropriately considered primates (Silcox, 2001). The body
mass estimates of P. petersonorum and the smallest Shanghuang primate dis-
cussed by Gebo et al. overlap, so it is unclear which species is actually the
smallest described primate.
Rose and Bown (1996) also included a previously described species of plesia-
dapiform, Alveojunctus minutus Bown, 1982, in the Picromomyidae on the basis
of similarities in the morphology of the enlarged P and the shared presence of
4
’ DepartmentofAnthropology, PennStateUniversity,409CarpenterBldg.,UniversityPark,PA 16802.
^Department of Cell Biology and Anatomy, Johns Hopkins School of Medicine, 725 N. Wolfe St.,
MD
Baltimore, 21205.
^Department ofPaleontology, San Diego Natural History Museum, P.O. Box 121390, San Diego, CA
92112.
Submitted 13 April 2001.
1
—
2 Annals of Carnegie Museum VOL. 71
low-crowned lower molars with shallow or absent hypoflexids, strongly mesially
canted trigonids, twinned entoconid and hypoconulid separated by an entoconid
notch, and weak or absent mesiobuccal cingulids. Hooker et ah (1999) challenged
the link between Picromomys and Alveojunctus, suggesting that the relationships
of the latter “will remain obscure until more material is found (p. 393).” These
authors did not account for the close similarity in P morphology seen in Picro-
4
momys and Alveojunctus, or for the molar characteristics shared by these two
genera. Although further material of Alveojunctus may suggest a different reso-
lution, the current evidence strongly supports the inclusion of both species in a
taxonomic grouping to the exclusion of any other known plesiadapiform.
Rose and Bown (1996) presented a cladistic analysis to elucidate the wider
relationships of the Picromomyidae. In spite of some general resemblances to
Picrodontidae (e.g., low-crowned molars and strongly mesially canted molar tri-
gonids), the results of the analysis suggested that these similarities are due to
convergence and that picromomyids are most closely related to either uintasori-
cine microsyopids or micromomyids. Although the authors slightly favored the
micromomyid tie, they conceded that formalizing this relationship taxonomically
was premature given the paucity of the material available at that time. Subse-
quently, Stafford and Szalay (2000) opined that the Picromomyidae were a group
of microsyopids, although this view was not supported by new specimens, a
revised cladistic analysis, or a discussion of any character evidence.
This paper documents new specimens of the Picromomyidae, including the
description of a new species ofAlveojunctus based on a single lower molar from
the Uintan of California. We recognize that naming a new species based on a
single tooth is ordinarily an undesirable practice. In this instance, however, we
deem it appropriate because of the extreme scarcity of this group, the diagnostic
differences between the new specimen and known picromomyids, and the tem-
poral and geographic separation between this specimen and the rest of the known
material of the Picromomyidae. The other species ofAlveojunctus, A. minutus, is
known from only a few isolated teeth, and its holotype is also a lower molar,
allowing the most appropriate comparisons to be made (Bown, 1982). To date,
only nine specimens that can be confidently referred to the Picromomyidae (see
below) have been discovered, seven of which are isolated teeth ofAlveojunctus.
This material is known from Wasatchian, Bridgerian, and Uintan deposits, indi-
cating that the family existed for at least six million years. In light ofthe derived
morphology exhibited by known picromomyids, compared to that of their likely
closest relatives, the family probably originated significantly earlier. Previous to
this report the geographic range ofthe Picromomyidae was limited to Wyoming
with the description here of a specimen from California the geographic range of
this family is significantly increased. All ofthese indications show that the family
is so rare that there can be no guarantee that any better specimens of the new
species will ever be recovered.
Dental characteristics of known picromomyids are suggestive of a diet similar
to that of the tiny extant marsupial feathertail glider, Acrobates (Rose and Bown,
1996), which feeds on a diet of insects, nectar, pollen, and possibly tree exudates
(Fleay, 1947; Woodside, 1995; Nowak, 1999), although further study is required
to rule out other possibilities (e.g., a diet of soft fruits).
A new specimen of Picromomys petersonorum that includes the first complete
lower central incisor is also documented here. The morphology of this tooth pro-
vides new evidence on the wider relationships of the family.
—
2002 SiLCOX ET AL. New Picromomyids 3
Abbreviations used in text are as follows: FMNH, Field Museum of Natural
History, Chicago, Illinois; SDSNH, San Diego Natural History Museum, San
Diego, California; USGS, United States Geological Survey, Denver, Colorado;
USNM, United States National Museum of Natural History, Smithsonian Insti-
tution, Washington D.C.
Systematic Paleontology
Order Primates Linnaeus, 1758
Suborder Plesiadapiformes Simons and Tattersall, in Simons, 1972
Family Picromomyidae Rose and Down, 1996
Genus Alveojunctus Bown, 1982
Alveojunctus bowni, new species
(Fig. lA)
— M
Holotype. SDSNH 31788, right Mj or Only known specimen; collected
2.
by M. A. B—oeder and S. L. Walsh in 1986.
Locality. San Diego Society of Natural History locality 3373, early Uintan,
upper tongue of the Friars Formation, Carmel Mountain Ranch housing devel-
opment, Unit—15, San Diego Co., California.
Diagnosis. M, ^^2 approximately 25-30% larger in linear dimensions than Mj oj.2
in Alveojunctus minutus. Lower molar paracristid less well defined than in A.
minutus. Strong, straight valley present between the protoconid and paraconid/
metaconid, missing in A. minutus. Contrasts with Picromomyspetersonorum (but
not A. minutus) in the lack of cingulids on the lower molar, the buccal cristid
obliqua resulting in the absence of a hypoflexid, the poor definition of the buccal
cusps, and the absence of a protocristid and postvallid.
—
Etymology. For Thomas M. Bown, in recognition of his valuable contributions to knowledge of
Paleogene mammals generally, and particularly to the discovery and study ofplesiadapiformprimates
including Picr—omomyidae.
Description. The single known isolated tooth cannot be definitively identified as M, or M2 (see
Discussion). The lack of a distal expansion associated with an enlarged hypoconulid, present on M3
of all plesiadapiforms except picrodontids, indicates that it is not likely to be M3. It is very low
crowned. The tooth has a straight mesial border, associated with a poorly demarcated paracristid and
a paraconid that is situated far lingually on the tooth, being positionedjust mesial to the metaconid.
The paraconid and metaconid, although closely appressed, are separated by a notch, so that each cusp
is distinct. A narrow talonid notch separates the metaconid and entoconid. These cusps are closely
positioned, compressing the talonid notch, and effectively closing the talonid basin lingually (in spite
ofthe absence ofan entocristid). The lingual cusps are better defined than the buccal cusps, although
allthecusps arelow,broad, andbulbous swellingsratherthan sharp,pyramidal structures. Thebroadly
swollen bases of the lingual cusps create a scalloped border to the central basin of the tooth (see
below).
A chip ofenamel is missing on the mesiobuccal margin ofthe protoconid, but the cusp appears to
have been broadly based and displaced mesially, so that it is almost directly buccal to the paraconid.
A faint postprotocristid extends down the distal face of the protoconid and merges with the cristid
obliqua, which is fully buccal in position. The buccal border ofthe tooth is convex (although this is
accentuated by the minordamage to the protoconid), so there is no sign ofan indentation (hypoflexid)
distinguishing the trigonid from the talonid. The presence or absence, or precise position, ofa hypo-
conulid is unclear due to slight chipping of the enamel. The lingual cant to the back of the tooth is
suggestive ofa lingually positioned hypoconulid, but this cannot be confirmed. Ifa hypoconulid was
present it must have been small, because the damage is very limited and there is no evidence ofsuch
a feature extending into the talonid basin.
There is no evidence ofa protocristid or postvallid, making the trigonid and talonid basins contin-
uous with one another. There is a straight, distinct valley between the protoconid and paraconid/
metaconid that extends this composite basin even further mesially, to the front of the tooth. These
4 Annals of Carnegie Museum VOL. 71
—
FBi.g.Al1v.eojAu.ncAtluvseomjiiniuntcutsu,s bUoSwGniS, 2n0e0w5s(pheoclioetsy,peS)D,SRNMH, 3127^8s8ter(heoolpohtoytpoeg)r,apRhM.,Mes2i^aslteisretoophtohteogtroappho.f
the figure and lingual to the left.
features make the crown of the tooth approximate a single, large, central basin surrounded on the
periphery by low cusps and crests.
—
Measurements (mm) of the Holotype. M, maximum width = 1.60 mm,
2
maximum length = 2.25 mm. Measured with an ocular micrometer to the nearest
0.05 mm. —
Discussion. SDSNH 31788 was tentatively assigned by Walsh (1996:85) to
the ?Palaeanodonta. Its revised identihcation here as pertaining to a new species
—
2002 SiLCOX ET AL. New Picromomyids 5
of Alveojunctus thus eliminates any record of palaeanodonts from the Uintan of
southern California.
SDSNH Loc. 3373 has yielded a diverse assemblage of small mammals of
early Uintan age. It was collected by screen-washing approximately 5000 kg of
m
bulk matrix from a 1.2 thick, caliche nodule-bearing, olive greenish gray silt-
stone paleosol. Other primate taxa represented at this locality include Omomys
sp. cf. O. carteri, Washakius woodringi, and Uintasorex montezumicus (see Lil-
legraven, 1980; Honey, 1990; Walsh, 1996). The stratigraphic and geographic
position of SDSNH Loc. 3373 was shown schematically by Walsh et al. (1996:
fig. 3). M
The identification of SDSNH 31788 as Mj or 2 is complicated by uncertainty
surrounding the identification of the holotype of Alveojunctus minutus, USGS
2005, This tooth was originally identified as an Mj (Bown, 1982), presumably on
the basis of the triangular trigonid that is relatively narrow compared to the tal-
M
onid. 2S in plesiadapiforms tend to Mhave trigonid and talonid regions that are
more nearly equal in width. Also, in 2 the trigonid is generally less triangular
than in Mj as a result of its mesiodistal compression, a flatter mesial border, and
a lingually positioned paraconid. These are also distinctions between USGS 2005
and the new specimen. In USGS 2005 there is a shallow hypoflexid, distinguishing
the trigonid from the talonid (see Fig. IB), that is missing in SDSNH 31788 (see
Fig. lA). Also, the flatter mesial border of SDSNH 31788 and more fully lingual
position of the paraconid relative to USGS 2005 are consistent with differences
between tooth positions (although there are some exceptions to these generalities,
as seen in the lower molars ofPurgatoriusjanisae and Palenochtha minor). Based
on tMhis reasoning it seems probable that USGS 2005 is an Mj and SDSNH 31788
an
2.
Rose and Bown (1996) suggested, however, that USGS 2005 is more likely
M
2, based on its similarity to that tooth in Picromomyspetersonorum. ThatMtaxon
has a trigonid and talonid of Mj that are subequal in width, whereas on the
2
trigonid is clearly narrower than the talonid. Picromomys also has a relatively
M
longer trigMonid on M, than on 2. In both of these features USGS 2005 is more
similar to of Picromomys than to M,. The straighter mesial border and more
2
nearly equal width of the trigonid and talonid in SDSNH 31788 may make this
more plausibly an Mj, based on the comparison with Picromomys. It is not clear,
M
however, whether the unusual pattern of differences between M, and in Pi-
2
cromomys stems solely from the different expression ofthe mesiobuccal accessory
cusp in this taxon. Particularly, the greater development of this cusp on Mj, rel-
M
ative to makes the mesial border of the tooth appear more squared off, and
2,
adds to the length and width of the trigonid. Since this cusp is missing in Alveo-
junctus it might be expectedMto show the more typical plesiadapiform pattern of
differences between Mj and 2. For these reasons, the attribution of the tooth is
left open until further material becomes available.
This discussion may prompt the criticism that the differences between USGS
2005 and SDSNH 31788 can be attributed to tooth position alone. The pronounced
size disparity between the specimens (we measured USGS 2005 as 1.80 mm long
and 1.20 mm wide; SDSNH 31788 is therefore 25% longer and 33% wider than
USGS 2005) argues against this. The size difference subst—antially exceeds that
generally seen between tooth positions in plesiadapiforms in Picromomys pe-
M M
tersonorum, for example, Mj is only 10% longer than 2, and 2 is slightly
wider (7%) than M,. This dental size disparity also translates into a significant
6 Annals of Carnegie Museum VOL. 71
difference in body size. If it is assumed that both molars are Mj, body mass
estimates for Alveojunctus minutus range from 50-111 g (Conroy, 1987: all pri-
mate or prosimian regressions = 50 g; Gingerich et aL, 1982: primate regression
= 111 g), whereas the range of estimates for Alveojunctus bowni suggests it may
have been twice the mass ofA. minutus (Conroy, 1987: all primate regression ==
124 g, prosimian regression = 114 g; Gingerich et ah, 1982: primate regression
= 2M38 g). These estimates do not differ significantly if one of the specimens is
an rather than an M^.
2
Additionally, the presence of the deep valley between the protoconid and para-
conid/metaconid in SDSNH 31788 (missing in USGS 2005), argues against the
view that both of these teeth belong to the same species. The presence or absence
of a valley between the protoconid and paraconid/metaconid is not a difference
that is seen between tooth positions, and this feature is missing in all related
forms. The differences between USGS 2005 and SDSNH 31788 seem to be com-
prehensible as a further expansion, in the latter, of the broad central basin that is
characteristic of Alveojunctus. In A. minutus this feature is formed because the
protocristid and postvallid are absent, allowing continuity between the trigonid
and talonid basins. In A. bowni, in addition to these features, there is a further
expansion forward created by the strong valley between the protoconid and para-
conid/metaconid. The reduction in the paracristid allows this valley to extend to
the front of the tooth. The mesial displacement of the protoconid in A. bowni
relative to A. minutus is probably also related, since this broadens the part of the
trigonid basin that is directly continuous with the talonid basin. In all, the differ-
ences between the two species can be explained as an accentuation of the basic
adaptive pattern that characterizes Alveojunctus, in which the molars became al-
most flat basins surrounded by weak bulbous cusps and low crests. The geolog-
ically younger age ofA. bowni is also consistent with this view.
Alveojunctus bowni extends both the known temporal and geographic range of
the Picromomyidae. To date, the genus Alveojunctus is definitively known only
from middle Bridgerian and early Uintan deposits. Bown (1982) suggested that
FMNH PM28689 might belong to A. minutus. This tooth, from the Cathedral
Bluff Tongue (early Bridgerian), was attributed to Niptomomys sp. by West and
M
Dawson (1973). West and Dawson identified the tooth as a right but the molar
M 3,
illustrated is a left 3. Unfortunately, the specimen could not be located for study
(and may be lost). The larger size of this tooth, relative to A. minutus, might
suggest that the specimen could belong to A. bowni. FMNH PM28689 is unusual
in lacking a paraconid, a feature that is present in both SDSNH 31788 and USGS
M
2005. The paraconid is absent on 3 of paromomyid plesiadapiforms, and it is
indistinct in Niptomomys doreenae (see Rose et ak, 1993:fig. 1). In Picromomys
and micromomyids known from serially associated dentitions (Tinimomys gray-
bulliensis [Rose et aL, 1993:fig. 2] and MicroMmomysfremdi [Fox, 1984:fig. 1]),
however, a paraconid is clearly demarcated on 3. Since we consid^iAlveojunctus
to be more closely related to Picromomys and micromomyids than to Niptomomys
M
or paromomyids, an indistinct paraconid would not be expected as a feature
3
of the genus. These indications suggest that this tooth may be attributable to a
taxon that is more closely related to Niptomomys (as originally indicated by West
and Dawson) than to Alveojunctus.
Stucky (1982, 1984) referred an additional specimen to Alveojunctus. This iso-
lated tooth, a right M], (UCM 44681) was not available for study. Stucky (1982)
notes that this tooth has more clearly delineated cusps and a higher trigonid than
—
2002 SiLCOX ET AL. New Picromomyids 7
Alveojunctus minutus. In light of the fact that Alveojunctus bowni differs from A.
minutus in precisely the opposite ways (i.e., it has a less well defined cusps and
a lower trigonid), it seems unlikely that this specimen could be referred to the
new species.
Picromomys petersonorum Rose and Bown, 1996
(Fig. 2)
—
A
Discussion. fragmentary left dentary with a nearly complete Ij, P4 talonid,
Ml, and alveoli for M2.3 (USNM 494979; Fig. 2) was collected in 1996 by M. T.
Silcox from Dorsey Quarry (U.S.G.S. locality D-2035Q; 397m above the base of
the Willwood Formation; see Bown et ah, 1994, and Silcox and Rose, 2001) in
the Bighorn Basin ofnorthwestern Wyoming. Although the Ii has a break between
the root and the rest of the tooth, and the crown is now displaced relative to the
dentary, it was found in continuity with the dentary and is clearly part of the
same specimen. The morphology of the P4 talonid and Mj matches that of the
holotype of Picromomys petersonorum.
The root of Ij is enlarged, laterally compressed, and is in a nearly horizontal
position in the mandible, as in the holotype (Rose and Bown, 1996). It bears a
distinct crest close to the lateral margin of the dorsal surface. Near the base of
the crown this dorsal crest crosses from its lateral position to the medial side of
the crown, where it descends onto the medial surface and runs distally to a point
approximately halfway from the base of the tooth, before fading away. Near the
point of this cross-over, another crest begins at the dorso-lateral extent of the
crown. The two crests present on the crown of the tooth define a surface that
exists in two planes, facing dorsally and medially.
Rose and Bown (1996) noted some basic differences between the morphology
of the lower central incisor preserved in the holotype of Picromomys petersono-
rum and that seen in microsyopids. Although the holotype preserves only a part
ofthe root ofIj, it is clear that Ij lacks the sharp dorsal border and basal expansion
characteristic of microsyopid incisors. Rose and Bown pointed out that the dis-
tinctive morphology of microsyopid incisors has resulted from a medial rotation
around the long axis of the tooth, so that the homolog of the lateral border of Ij
in most plesiadapiforms forms a sharp dorsal crest in microsyopids. Rose and
Bown (1996) argued that such a rotation had not occurred in Ii ofpicromomyids.
It remained possible, however, that when the rest of the crown became known it
would be more microsyopid-like, or else represent a recognizable intermediate
stage between the morphology observed in other plesiadapiforms and microsy-
opids.
The morphology ofUSNM 494979 does not support this view. The dorsal crest
noted by Rose and Bown (1996) on the root is continuous with the medial crest
on the crown of the Ij in USNM 494979. In microsyopids, however, the dorsal
crest is homologous with the lateral crest on the crowns of more typical plesia-
dapiforms (e.g., Plesiolestesproblematicus; see Bown and Gingerich, 1973). This
indicates that the Ii morphology in Picromomys is not an intermediate on the path
to the microsyopid condition. Furthermore, the shape of this incisor differs mark-
edly from that observed in microsyopids in lacking a basal expansion of the
crown. Also, the portion of the crown that is delimited by crests is not flat, as in
microsyopids, but incorporates two distinct planes. In all, the picromomyid I,
morphology is less consistent with a relationship to microsyopids than that ob-
8 Annals of Carnegie Museum VOL. 71
view
buccal
oblique
in
2.3
M
for
alveoli
and
M,,
P4,
of
talonid
I,,
containing
dentary
left
Incomplete mm.
1
=
bar
494979.
Scale
USNM
(above).
Ij
of
petersonorum,
view
occlusal
Picromomys
with
—
2.
(below),
Fig.
P
—
2002 SiLCOX ET AL. New Picromomyids 9
served in some palaechthonids. Even some primitive plesiadapoids have a more
closely comparable incisor morphology (e.g., Elphidotarsius sp. cf. florencae, in
which I, is broad at the base but not rotated medially; see Rose, 1975), in spite
of the fact that plesiadapoids are not very closely related to microsyopids (Silcox,
2001
).
The Ij of Picromomys is similar to this tooth in a number of plesiadapiforms,
including Saxonella, Picrodus, paromomyids, micromomyids, and possibly PaP
enochtha minor (see Gunnell, 1989:fig. 5), in being slender, laterally compressed,
and elongate. The evidence available for Purgatorius is also consistent with the
presence of this morphology, although the tooth itself is still unknown for this
basal plesiadapiform (Silcox, 2001). The widespread distribution of this basic
morphology suggests that it may be primitive for plesiadapiforms (Silcox, 2001).
The lower incisor of Picromomys is particularly similar to that of the only mi-
cromomyid for which this tooth has been published, Tinimomys graybulliensis
(i.e., USNM 425583, Beard and Houde, 1989). As in Picromomys, Ij in Tini-
momys shows a similar reorientation of a dorsolateral crest in the root region to
a medial position on the crown. This change in direction in the dorsal crest is not
present in any of the other plesiadapiforms examined. When added as a character
to the dataset published by Rose and Bown (1996) (see Appendix), an exhaustive
search using PAUP* 4.0p6 (Swofford, 2001) resulted in a single most parsimo-
nious tree corresponding to their Tree B (length = 73 steps. Cl = .49, RI = .537,
statistics from PAUP*; see Rose and Bown, 1996:fig. 5). This tree supports a
clade including Picromomys, Alveojunctus, and Tinimomys, to the exclusion of
Niptomomys, and portrays picromomyids as most closely related to micromo-
myids. —
Measurements (mm) of USNM 494979. width = 0.80 mm; Mj trigonid
4
width = 0.70 mm, Mj talonid width = 0.80 mm, M, length = 1.20 mm. Measured
with an ocular micrometer to the nearest 0.05 mm.
Conclusions
This paper documents the eighth and ninth confidently attributed specimens of
the Picromomyidae. The type specimen of the new species Alveojunctus bowni
from the early Uintan of California extends the temporal and geographic range
of the family. This specimen demonstrates an even more extreme version of the
basic pattern of picromomyid dental structure, leading to an extraordinarily flat
and simplified lower molar morphology. This is consistent with an increasingly
specialized diet. The second recognized specimen of Picromomys petersonorum
provides new information on the incisor morphology and suggests a stronger tie
to the Micromomyidae than to the Microsyopidae. This may imply that Micro-
momyidae is a paraphyletic group (Hooker et ah, 1999). Specifically, picromo-
myids and Tinimomys may be sister taxa to the exclusion of Micromomys (as
indicated in Tree B of Rose and Bown, 1996:fig. 5). In light of this, it may
eventually be appropriate to make Picromomyidae a subfamily of the Micromo-
myidae. Making definitive conclusions about the relationships of micromomyids
and picromomyids is hampered, however, by the very limited information avail-
able for most species of Micromomys (Beard and Houde, 1989), and the lack of
specimens from the upper dentition of Picromomyidae. Consequently, we prefer
to maintain Picromomyidae and Micromomyidae as distinct families, pending
further discoveries or additional analyses.
10 Annals of Carnegie Museum VOL. 71
Acknowledgments
Figure 1 was taken at the Johns Hopkins Microscope Facility with supplies generously providedby
M. F. Teaford. Figure 2 was expertly prepared by Elaine Kasmer. Fieldwork in the Bighorn Basin of
Wyoming was carried out under BLM permits and supported by NSF grant IBN-9419776 to KDR.
Comparative studies of plesiadapiforms by MTS were supported by the Paleobiological Fund, the
Wenner-Gren Foundation for Anthropological Research, Sigma Xi, and NSF Doctoral Dissertation
Improvement Grant 9815884. Collection of SDSNH 31788 was made possible by Carmel Mountain
Ranch. This paper was significantly improved by the comments ofthree anonymous reviewers.
Literature Cited
Beard, K. C.,andR Houde. 1989. Anunusual assemblageofdiminutiveplesiadapiforms(Mammalia,
?Primates) from the early Eocene of the Clark’s Fork Basin, Wyoming. Journal of Vertebrate
Paleontology, 9:388-399.
Bown, T. M. 1982. Geology, paleontology, and correlation ofEocene volcaniclastic rocks, southeast
Absaroka Range, Hot Springs County, Wyoming. United States Geological Survey, Professional
Paper No. 1201-A:A1-A75, Washington, D. C.
Bown, T. M., and P. D. Gingerich. 1973. The Paleocene primate Plesiolestes and the origin ofthe
Microsyopidae. Folia Primatologica, 19:1-8.
Bown, T. M., K. D. Rose, E. L. Simons, and S. L. Wing. 1994. Distribution and stratigraphic
correlation of upper Paleocene and lower Eocene fossil mammal and plant localities ofthe Fort
Union, Willwood, and Tatman formations, southern Bighorn Basin, Wyoming. United StatesGeo-
logical Survey Professional Paper No. 1540:1-103, Washington, D. C.
Conroy, G. C. 1987. Problems ofbody-weight estimation in fossil primates. International Journal of
Primatology, 8:115-137.
Fleay, D. 1947. Gliders of the gum trees: the most beautiful and enchanting Australian marsupials.
Bread and Cheese Club, Melbourne, Australia.
Fox, R. C. 1984. The dentition and relationships ofthe Paleocene primateMicromomys Szalay, with
description ofa new species. Canadian Journal ofEarth Science, 21:1262-1267.
Gebo, D. L., M. Dagosto, K. C. Beard, and T. Qi. 2000. The smallest primates. Journal ofHuman
Evolution, 38:585-594.
Gingerich, P. D., B. H. Smith, and K. Rosenberg. 1982. Allometric scaling in the dentition of
primates and prediction ofbody weight from tooth size in fossils. American Journal ofPhysical
Anthropology, 58:81-100.
Gunnell, G. F. 1989. Evolutionary history ofMicrosyopoidea (Mammalia, ?Primates) and the rela-
tionship between Plesiadapiformes and Primates. University ofMichigan Papers on Paleontology,
27:1-157.
Honey, J. G. 1990. New washakiin primates (Omomyidae) from the Eocene ofWyoming and Col-
orado, and comments on the evolution ofthe Washakiini. Journal ofVertebrate Paleontology, 10:
206-221.
Hooker, J. J., D. E. Russell, andA. Phelizon. 1999. A new family ofPlesiadapiformes(Mammalia)
from the old world lower Paleogene. Palaeontology, 42:377-407.
Lillegraven, j. a. 1980. Primates from later Eocene rocks ofsouthern California. Journal ofMam-
malogy, 61:181-204.
Linnaeus, C. 1758. Systemanaturaeperregnatrianaturae, secundumclasses,ordines,genera,species,
cum characteribus, differentiis, synonymis, locis. Vol. 1: Regnum animale. Editio decima, refor-
mata. Laurentii Salvii, Stockholm, Sweden.
Nowak, R. M. 1999. Walker’s Mammals oftheWorld, SixthEdition. Johns HopkinsUniversityPress,
Baltimore, Maryland.
Rose, K. D. 1975. The Carpolestidae: early Tertiary primates from North America. Bulletin of the
Museum ofComparative Zoology, 147:1-74.
Rose, K. D., and T. M. Bown. 1996. A new plesiadapiform (Mammalia: Plesiadapiformes) from the
early Eocene ofthe Bighorn Basin, Wyoming. Annals ofCarnegie Museum, 65:305-321.
Rose, K. D., K. C. Beard, and P. Houde. 1993. Exceptional new dentitions of the diminutive ple-
siadapiforms Tinimomys and Niptomomys (Mammalia), with comments on the upper incisors of
Plesiadapiformes. Annals ofCarnegie Museum, 62:351-361.
SiLCOX, M. T. 2001. A phylogenetic analysis ofPlesiadapiformes and theirrelationship toEuprimates
and other archontans. Unpublished Ph.D. Dissert., Johns Hopkins University, Baltimore, Mary-
land.
SiLCOX, M. T, and K. D. Rose. 2001. Unusual vertebrate microfaunas from the WillwoodFormation,